[Tinyos-devel] TelosB as base station?

[Matt Welsh]

Joe, good point - though I'm not sure that's the problem... the mote
seems to freeze up very quickly after it is restarted; I have not been
able to correlate this with a particular kind of traffic pattern (but
perhaps we should try). 

On Thu, 2007-07-05 at 11:50 -0700, Joe Polastre wrote:
> Does the problem occur immediately or after some time?  From what I
> recall, the DCO isn't being continuously recalibrated in the TinyOS
> 2.x release.  Temperature changes as small as 10 degrees can adversely
> affect the mote (ie, the exhaust fan on your laptop turning on and
> blowing hot air at your mote).  Also, the DCO drifts over time, and
> can be significant.
> 
> Just an FYI.
> 
> -Joe
> 
> On 7/5/07, Matt Welsh <mdw at eecs.harvard.edu> wrote:
> > Hi folks,
> >
> > We have been having continued problems with using a TelosB mote as a
> > base station and I'm wondering if there's something we're doing wrong.
> > Basically, using BaseStationCC2420 on a TelosB along with the C-based
> > serialforwarder is just incredibly unstable. It seems to hang after
> > bursts of packets and we usually have to restart the base station mote.
> > (The code in support/sdk/c/serialsource.c also prints the quite useful
> > error message 'foo' - apparently when no data is available on the serial
> > port - we commented that out long ago.)
> >
> > Not sure if this is a problem with the TelosB, the sf, or both. I
> > remember having similar problems with TinyOS 1.x and am somewhat
> > surprised this hasn't been resolved in 2.x - or maybe it has and we
> > don't know the fix?
> >
> > We do find that MicaZ motes are very stable as base stations (running
> > for weeks at a time without any hang).
> >
> > Thanks -
> > Matt
> >
> >
> >
> > _______________________________________________
> > Tinyos-devel mailing list
> > Tinyos-devel at Millennium.Berkeley.EDU
> > https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-devel
> >